def sentence_gen():
histo_text = get_words('siddhartha.txt')
histo = histogram(histo_text)
random_word = sample(histo)
random_words = []
for i in range(7):
random_words.append(sample(histo))
random_sentence = sentence_maker(random_words)
return random_sentence
max_freq = max(word_dict.values())
rand_frequency = random.uniform(0, max_freq)
list_from_words = list(word_dict)
while True:
rand_index = random.randint(0, len(word_dict) - 1)
selected_word = list_from_words[rand_index]
if word_dict[selected_word] >= rand_frequency:
return selected_word
def make_sentence(chain, starting_words, sentence_len):
previous_words = starting_words
selected_list = [starting_words[0], starting_words[1]]
for _ in range(sentence_len - 2):
selected_word = next_word(chain, previous_words)
if selected_word is not None:
selected_list.append(selected_word)
previous_words = (previous_words[1], selected_word)
sentence = ' '.join(selected_list)
return sentence
if __name__ == '__main__':
word_list = get_words('life.txt')
chain = markov_chain(word_list)
list_from_chain = list(chain)
random_words = random.choice(list_from_chain)
# print(random_words)
sentence = make_sentence(chain, random_words, 5)
print(sentence)
import random
from dictogram import Dictogram
from histogram import get_words
from stochastics import sampler
corpus = get_words("corpus/corpus-0.txt")
def first_order(corpus_of_words):
words = corpus_of_words
chain = {}
corpus_length = len(corpus_of_words)
for i, key in enumerate(words):
if corpus_length > (i + 1):
word = words[i + 1]
if key not in chain:
chain[key] = Dictogram([word])
else:
chain[key].add_count(word)
return chain
def second_order(corpus_of_words):
words = corpus_of_words # list of strings
chain = {} # dict to hold Markov states, key: word, value: histogram
corpus_length = len(words)
for i, word1 in enumerate(words):
if i + 2 >= corpus_length:
return self[tokens].sample()
def build_sentence(self, num_words, words_list):
self.build_state_histogram(words_list)
sentence = []
first_words = random.choice(list(self.keys()))
tokens = Queue(first_words)
sentence.extend(first_words)
total_words = len(first_words)
while total_words < num_words:
next_word = self.get_next_word(tuple(tokens.items()))
if next_word == '**STOP**':
sentence.append(next_word)
break
sentence.append(next_word)
tokens.dequeue()
tokens.enqueue(next_word)
total_words += 1
return ' '.join(sentence)
if __name__ == '__main__':
text = 'text/three_wishes.txt'
words_list = histogram.get_words(text)
markov = MarkovChain(4)
# markov.build_state_histogram()
# for key in markov:
# print(key, markov[key])
print(markov.build_sentence(6, words_list))
total_freq += index[1]
print(total_freq)
random_num = random.uniform(0, 1)
#returns random sample words
for index in listogram:
chance += index[1] / total_freq
if chance >= random_num:
return index[0]
if __name__ == '__main__':
#using histogram functions to get corpus
histo_text = get_words('fish.txt')
histo = histogram(histo_text)
test = total_freq(histo)
print(test)
#sampling using dictionary method
sample_word = sample(histo)
probability = print_probability(histo)
print(sample_word)
#testing that sampling function actually works
# outcomes = test_probability(histo)
# results = results_histogram(outcomes)
# print(results)
#list implementation of sampling
def random_word(source_text):
words_list = histogram.get_words(source_text)
return random.choice(words_list)
def sample(words_list):
word_distribution = get_word_distribution(words_list)
ran_num = random.randint(1, len(words_list))
for word in word_distribution:
if ran_num in word_distribution[word]:
return word
def test(words_list):
test = []
histogram_ = histogram.histogram_dict(words_list)
for _ in range(10000):
test.append(sample(words_list))
hist_ran_words = histogram.histogram_dict(test)
for word in hist_ran_words:
print(word, 'test', hist_ran_words[word]/10000,
'distribution', round((histogram_[word]/len(words_list)), 4))
return
# params = sys.argv[1:]
# words_list = histogram.get_words(params[0])
words_list = histogram.get_words('text/three_wishes.txt')
print(sample(words_list))
# print(get_word_distribution(words_list))
# test(words_list)
dictionary = words
if type(words) == list():
for word in words:
dictionary = {}
dictionary[word] = words.count(word)
random_value = random.random()
total_value = sum(dictionary.values())
total = 0
# number_wins = {}
for key, value in dictionary.items():
total += value
if total / total_value >= random_value:
return key
# if key in number_wins:
# number_wins[key] += 1
# else:
# number_wins[key] = 1
# break
if __name__ == '__main__':
words = get_words('animals.txt')
histogram_dict = get_words_counts_dict(words)
# sampler(words)
print(test_sampler(words, 10000))